Cognizant 20-20 Insights

An Analytical Approach to Provider

and Intermediary Segmentation
in the Pharmaceuticals Industry
Given the influence of managed markets, many pharmaceuticals
are aggressively targeting providers and intermediaries such as
pharmacy benefit management groups and health plan payers. Using
statistical techniques, pharmas can effectively segment customers
to design market-driven strategy, launch new therapies, market
existing products and maximize the value of in-line brands.

Executive Summary statistical rigor required, ease of communica-

The economic environment for pharmaceuticals
companies is growing increasingly competitive.
Understanding the best ways to apply market
segmentation can provide a variety of benefits
for a forward-thinking company. Well-designed
segmentation schemes can increase pharma
companies understanding of customers and
their varying needs.
tion, granularity of insight, etc. These are critical
for marketing the right products to the right
customers at their time of need.
This white paper highlights the key elements
and guiding principles that will help sales and
marketing teams design customer segmentation
strategies to target key influencers, including:

From our engagement experience with pharma

companies, it is clear that many struggle to make
informed decisions to maximize their focus on
the right sets of customers and segments in
order to boost revenue. In our experience we
have found that employing multiple segmen-
tation techniques along the insight-complex-
ity spectrum can help alleviate this struggle.
Identification of an appropriate segmentation
technique is based on multiple factors such as
nature, type, usability of data/variables, level of
Providers: National/regional hospitals, medical
groups and integrated delivery networks.
Intermediaries:Pharmacy benefit managers
(PBMs) and payers.
A Market Segmentation Primer
As markets have become more competitive,
establishing a right segmentation strategy is
critical for achieving market potential. However,
most companies still struggle to deliver market
insights that create relevant, actionable and

cognizant 20-20 insights | february 2017

Traditional Market Segmentation Challenges

Product launches
are not taken into
account
End customers
(patients) needs
are not
captured

Challenges
Lack of
objective-based
segmentation based
on robust analytical
framework to
identify relevant Too many priority
segments customers

Figure 1

differential segments that take into account Applying an Analytics Approach

product portfolio needs and future business
realities. Considering these challenges, orga-
nizations require an analytical approach to
identify key drivers that impact market potential
while ensuring future portfolio needs (e.g., the
product pipeline) are captured and eventually
addressed. Earlier research studies in the phar-
maceuticals industry primarily cover physician
and hospital segmentation approaches. This
white paper covers segmentation approaches
that are especially applicable to providers and
intermediaries. Figure 1 depicts the current
business situation.
to Differential Segmentation for
Providers and Intermediaries
Figure 2 depicts an analytics-driven approach
that uses objective and subjective factors for
arriving at a robust segmentation scheme.
The chronological steps for segmentation are as
follows.
1. Build Analysis Universe
Customers are defined for related
brands or product portfolio/thera-
peutic area. If segmentation strategy
is not specific to any particular brand, then all
such entities need to be accounted for that are
related to the pharmas portfolio/therapy. These
entities should have impact on the companys
market potential (including competitors sales).

Documenting Analytics Process Flow

Build Variables Identify Key Drivers Segmentation

Analysis Identification to Arrive at Using Clustering
Universe and Computation Differential Segments Techniques
Figure 2

cognizant 20-20 insights 2

 Providers: When starting the segmentation
exercise, its important to understand the rela-
tionship among integrated delivery networks
(IDNs), hospitals, group practices and clinics,
as some of these entities could be possible
affiliates of other entities. In other words, its
better to narrow down the analysis to parent
entities, considering their influence and control
over their affiliates.
Intermediaries: Not all the intermediaries
in the managed care market are relevant
to enhance a drugs revenue. For example,
certain intermediaries limit their formulary
for generic drugs only, some might deal in a
specific therapeutic area, while others may not
offer sufficient patient benefits to accelerate
the sales of an expensive drug. Based on the
specific characteristics of intermediaries
namely, pharmacy benefit managers (PBMs),
payers or group purchasing organizations
(GPOs) its highly beneficial to narrow the
analysis universe to a condensed set of entities
of interest to design a segmentation scheme
with reduced effort.
2. Variables identification
This stage focuses on the formation
of multifold independent variables
that can alter and best explain
variations in brand sales.
Key objective factors/variables include:
Provider:
>> Entities size (in terms of physicians, staffed
beds, etc.) and contribution to brand sales.
>> Entities sophistication level.
>> Entities that emphasize patient and clinical
care.
>> Entities that exert high control/influence.
>> Entities that would contribute to future
pipeline drugs.
Institutions that contribute to greater brand
share compared with other institutions within the
same geography tend to exert higher influence
and dominance. In fact, such institutions need to
be continuously promoted and engaged to ward
off competitors.
Intermediary:
>> Entities size (in terms of lives covered) and
contribution to brand sales.
cognizant 20-20 insights
>>Entities that have preferred formulary
access with no restriction.
> >Entitiescontribution to commercial,
Medicaid and Medicare channels.
> >Entities
that exert high geographical
dominance.
>> Entities contribution to branded or generic
drugs.
>> Entitiesassociated with plans that have
strong ratings in terms of patient treatment
experience.
Both secondary and primary data sources are
leveraged for identifying these variables.
Secondary data sources:
>>Plan level data provides sales information
by product and by market at the plan and
payer levels. This is used by pharma compa-
nies to capture Rx and sales information for
providers and intermediaries.
>>Hospital level data provides information
on hospitals, medical groups and measures
such as the number of physicians, geograph-
ical reach, staffed beds, admissions, elec-
tronic medical records, e-prescribing, etc.
>>Formulary data indicates how many
lives are covered in various plans, and
it can be further rolled up to the payer
level. It also captures payer/product tier
information.
Pr im ar y d ata sou rces for h osp ital
performance:
>>Centers for Medicare and Medicaid Services
(CMS) compares hospital performance in
patient care.
>>Industry standard benchmark hospital
surveys, such as the American Hospital
Associations most-wired hospital survey
which indicates the level of IT adoption in
U.S. hospitals and health systems.
>>HIMSS tracks healthcare providers prog-
ress toward paperless patient record envi-
ronment adoption (e.g., electronic medical
records).
>>Institutionalized ratings/market research
assessments.
>>PayCo Score is a composite measure of
risk associated with not contracting with
a given payer.
3
 3. Key Driver Analysis
The purpose of key driver analysis is
to figure out important metrics that
can impact brand sales and possibly
market potential. The outputs of this stage are
the key components/metrics that are uncor-
related and explain maximum variability in the
data set. In order to determine the key drivers,
pharma organizations must conduct:
Correlation checks among independent
variables/metrics before any clustering
analysis, to avoid interaction effects.
Principal component analysis (PCA), which
deals with the problem of correlating too many
variables. Key drivers can be identified that
explains ~80% of the variability in the data set.
Another approach is to conduct regression
analysis, where a dependent variable could
be brand sales or market potential (including
competitor sales). Note: To achieve proper
segmentation, it is very helpful to identify the
significant independent variables that impact
brand sales or market potential.
Resultant key drivers are used in the next stage:
setting segmentation strategy.
4. Segmentation Design Using
Advanced Analytical Techniques
Conducting customer segmenta-
tion can follow several different
approaches, which vary from simple decile-based
segmentation to complex clustering techniques.
Figure 3 depicts four advanced analytical
techniques that are popular and widely used for
customer segmentation.

Segmentation Techniques: A Comparative View

Segmentation Approach Pros Cons

Factor Analysis Segmentation Can address the problem of Might fail to capture
having too many variables to consumers multifaceted
correlate. nature.

Relatively simple to execute.

K-Means Clustering Overcomes the pitfalls of
factor analysis by providing
information about each
attributes contribution to
differentiating the clusters.
Can take individual attributes
or factors as inputs.
Not recommended for cat-
egorical, ranking variables.
Preclustering treatment
needs to be done for consid-
ering uncorrelated and stan-
dardized/scaling variables.

Faster than the below

approaches.

Hierarchical Clustering Can use variables with Entities with any missing
differing scale types including values are excluded from the
categorical variables. analysis.

Offers guidance about the Has slow execution time, so

optimal number of clusters works best for smaller data
(whereas factor segmentation sets.
and k-means do not).

Latent Class Analysis (LCA)
Since it is based on probability Can take longer to run versus
modeling, segments are more other approaches.
likely to be real as compared
with the above approaches. Continuous variables need to
be categorized.
Preferable for multivariate
categorical data.

Scaling/standardization of
data is not a prerequisite.

Figure 3

cognizant 20-20 insights 4

 Quick Take
Identifying Priority Targets by Segmenting Managed Markets Customers
We helped one of the top 10 global pharma companies to identify targets for segmentation of managed
market team customers: providers and intermediaries (PBMs and payers) in North America. The client
therapeutic focus is on cardiovascular, oncology, respiratory and neuroscience areas.

Challenges in Existing Segmentation

Need to be based on metrics that are objective and tangible.
Future portfolio needs (product pipelines) are not captured in existing segmentation.
Key drivers that impact market potential need to be determined.
Higher number of targets (priority customers).
Need for customer-centric engagement to ensure a replicable and transparent
segmentation process.

Approach

Variable Identification Pre-clustering Analysis Clustering & Identifying Priority

& Profiling to Identify Key Drivers Segmentation Targets

Variables were derived/ Correlation checks K-Means clustering Prioritize customers

identified from different
secondary sources.
Specific metrics such
as percent of specialty
affiliates, owned health
plans and control are
included as objective
factors.
executed.
Principal Component
Analysis (PCA) used to
obtain uncorrelated
& transformed key
variables.
technique used for considering:
determining clusters.
>> Contracting: sales,
Clusters were profiled control, etc.
and analyzed to create
actionable segments. >> Brand strategy: brand
share in market.
>> Resourcing: size and
nature of an entity.

We analyzed some of the below metrics for segmentation of customer groups providers and
intermediaries (PBMs & payers):

Intermediaries Providers

Size: e.g., lives covered, gross sales, revenue etc.
PayCO Score
Control: e.g., percent brand Rx share
within geography
Percent Share by Payment Type: e.g., prescription,
units sold, sales by payment type (Medicaid, Medicare,
etc.)
Formulary Access: key brands assigned to the
corresponding formulary position
NCQA Plan Ratings
Size: e.g., total physicians, staffed beds, admissions,
affiliates, gross sales
Control: e.g., percent brand Rx share
within geography
Own Health Plans: provider sponsored & enrollments
Quality: e.g., adherence to CMS guidelines, patient
satisfaction
Sophistication/Technology: e.g., use of e-prescribing,
utilization of EMR, sales rep access, formulary status
Percent Specialty Affiliates: related to impacted
therapy in future portfolio

Key Benefits

Actionable segments that meet current/future portfolio needs were determined.

Segments profiled were straightforward to communicate e.g., focused Innovators, influential
pacesetters, etc.

Narrowed field to priority customers for providers and intermediaries that are most receptive and
valuable.

cognizant 20-20 insights 5

Looking Ahead But not every segment is lucrative for an orga-
nization. Considering this, pharmaceuticals
Analytical approaches to market segmenta-
companies can begin the targeting journey by
tion can help pharma companies to identify
prioritizing the customers within a segment.
relevant, actionable and measurable target
Priorities for targeting would differ from one
customers among providers and intermediaries.
organization to another, depending on business
Each segment is typically defined and named
objectives. The key attributes to consider for
based on its characteristics. These segments
customer prioritization after segmentation
are given straightforward names that commu-
include:
nicate meaningful insights to field personnel.
Characteristics such as percent of brand share,
number of entities within segment, etc. also
Contracting: Based on sales/size, influence/
control, etc.
reveal segment attractiveness and value.
Brand strategy: Based on brand share in
market.

Resourcing: Size and nature of an entity.

The common element for success is the use
of a disciplined and consultative approach to
segmentation.

References
www.pharmexec.com/brand-strategies-and-growing-influence-idns

www.hhnmostwired.com/

cognizant 20-20 insights 6

About the Authors
Payal Mangal is a Manager within Cognizants Life Sciences Analytics and Information Management
Practice. She has over 10 years of market research and analytics experience in the life sciences domain.
Payal has extensive experience in sales and marketing analytics focusing on call planning, alignment,
customer segmentation, brand analytics dashboards and market research, as well as a focus on business
development for global life sciences clients. She holds a masters degree in information technology.
Payal may be reached at Payal-1.Mangal-1@cognizant.com.

Gaurav Bhardwaj is a Senior Associate within Cognizants Life Sciences Analytics and Information
Management Practice. He has over seven years of experience in the life sciences domain. Gaurav has
rich experience in marketing analytics projects such as customer segmentation and campaign ROI for
major pharmaceuticals clients. He has also helped clients in sales force alignment design, personnel
placement and sales force skill assessment projects. Gaurav has an M.B.A. from Great Lakes Institute of
Management, India, and a B.E. from Netaji Subhas Institute of Technology, University of Delhi. He can be
reached at Gaurav.Bhardwaj2@cognizant.com.

Asif Rizwan is a Senior Associate within Cognizants Life Sciences Analytics and Information Management
Practice. He has helped clients in sales force alignment design, segmentation, forecasting and sales
force skill assessment projects. Asif holds an M.E. from IIT, Madras. He can be reached at Asif.Rizwan@
cognizant.com.

About Cognizant
Cognizant (NASDAQ-100: CTSH) is one of the worlds leading professional services companies, transforming clients
business, operating and technology models for the digital era. Our unique industry-based, consultative approach
helps clients envision, build and run more innovative and efficient businesses. Headquartered in the U.S., Cognizant is
ranked 230 on the Fortune 500 and is consistently listed among the most admired companies in the world. Learn how
Cognizant helps clients lead with digital at www.cognizant.com or follow us @Cognizant.

World Headquarters European Headquarters India Operations Headquarters

500 Frank W. Burr Blvd.
Teaneck, NJ 07666 USA
Phone: +1 201 801 0233
Fax: +1 201 801 0243
Toll Free: +1 888 937 3277
Email: inquiry@cognizant.com
1 Kingdom Street
Paddington Central
London W2 6BD
Phone: +44 (0) 20 7297 7600
Fax: +44 (0) 20 7121 0102
Email: infouk@cognizant.com
#5/535, Old Mahabalipuram Road
Okkiyam Pettai, Thoraipakkam
Chennai, 600 096 India
Phone: +91 (0) 44 4209 6000
Fax: +91 (0) 44 4209 6060
Email: inquiryindia@cognizant.com

Copyright 2017, Cognizant. All rights reserved. No part of this document may be reproduced, stored in a retrieval system, transmitted in any form or by any
means, electronic, mechanical, photocopying, recording, or otherwise, without the express written permission from Cognizant. The information contained herein is
subject to change without notice. All other trademarks mentioned herein are the property of their respective owners.
Codex 2303